The long-range goal of this project is to develop a non-human primate model for studies of therapy for human cytomegalovirus (HCMV) that will 1) enable rapid in vivo evaluation of promising anti-HCMV drugs, and 2) allow development of therapeutic strategies for HCMV under clinically relevant conditions. Rhesus CMV (RhCMV) infection of rhesus macaques was chosen for this project because, like HCMV in humans, RhCMV typically establishes lifelong persistent, but asymptomatic, infections in healthy individuals, but it causes substantial morbidity in macaques with an impaired immune system. Preliminary studies have demonstrated that RhCMV and HCMV are nearly identical in in-vitro susceptibilities to approved anti-CMV drugs and to several members of a promising new class, benzimidazole ribonucleosides (BR). We hypothesize that RhCMV infection of rhesus macaques can be used for rapid and sensitive in vivo evaluation of anti-CMV drugs by assessment of changes in parameters of acute and persistent infection of healthy, immunocompetent macaques. We propose to evaluate the model with one well-characterized and approved anti-CMV drug, Cidofovir (CDV), and to characterize the anti-CMV activity of a promising BR. In Aim 1 we will quantify the reduction in RhCMV genome copy number in plasma by real-time PCR during primary infection in monkeys treated with either CDV or a BR, compared to controls. Another measure of the efficacy of drug treatment will be provided by analysis of host anti-CMV immune responses. In Aim 2 the efficacy of CDV and BR during persistent RhCMV infection will be quantified by reduction of the frequency and titer of RhCMV shed at the oral and genital mucosa. Ultimately, we predict that this model can be used to test therapies for CMV infections during AIDS, Transplantation, or fetal development in a primate host that can be experimentally manipulated.